Tuesday, November 9, 2010

Grand designs

So Hawking's new book is a bit of a disappointment. It boils down to the world is the way it is because it's the way it is; small things can't blow up but big things can and intelligence could spontaneously arise in the Game of Life so from little rules big things can grow.

A lot of what he says is reasonable, of course, and a lot is indisputable. But a signal problem I think he has is that while he does say that we do not have a description of reality, which we cannot achieve, but rather a model that we can investigate to see whether it coincides with reality, he then uses an interpretation of that model to be prescriptive about how the universe must be. What I mean is, we can model the behaviour of electrons using Feynman's sum over histories, and then look at the world to see whether it fits. The world does fit, but that doesn't confirm Feynman, because there are other models that also fit (also, we have to use renormalisation because in fact our mathematics doesn't work--Hawking seems entirely unconcerned that renormalisation only works for the electroweak force so we don't have a full theory of other forces).

So Hawking assumes Feynman's interpretation and draws a conclusion that pretty much wishes away all the questions he asks: why is there a universe? why is it like this? why is it so perfectly adjusted for us? He does this by suggesting that the strong anthropic principle holds but goes a bit further. It's fairly obvious that the universe has to be the way it is for us to be here to comment on it (the anthropic principle) but it has been assumed in the past that the laws of physics are the only ones possible. Hawking says no: there are many universes with different laws, and we happen to be in this one because no matter how probable other laws are, we can only be in universes with histories that lead to us.

This is supposed to disprove the existence of God. I know what you're thinking. God took a great deal of getting rid of if we needed a quadrillion universes to make him impossible. And that's a lot of universes to hang on the double slit experiment (versions of which are the basic experimental evidence for all this). I mean, it's not unreasonable to have your doubts that electrons really do take every conceivable path to their target. It involves some huge mathematical jiggery pokery to go from that to the observation that if you fire an electron from A to B, it ends up at B and not on Mars. The standard answer is that the straight line, and paths near it, is much more probable than going to Mars, which is fair enough, but we are yet to observe an electron straying. Not everyone agreed with Feynman: I think it was Bohm who argued that the electron takes a definite path and quantum theory simply describes our inability to measure precisely what it is. Not everyone believes an observer collapses the wavefunction. It's not at all impossible that the wavefunction describes a possibility but electrons do take only one path.

I'd certainly expect there to be in time an explanation of the nonlocal effects of quantum theory that didn't look so much like magic. Quantum entanglement is the kind of thing that is key for Hawking's thesis. In QE a pair of particles is formed that has opposite values of a certain parameter. One is positive, the other negative. They are separated and taken to different ends of the lab, so that they would need a measurable time to signal their state to each other. Then the polarity of one is changed, and the other instantly changes. Hawking will say that this is achieved by the necessity of the observer being in a universe in which the polarity of the second particle has changed. I am not sure that there's much mileage for science in saying no more than it's how it is because it has to be like that.

I didn't really understand his explanation of how the universe came into being. I have to confess, my eyes glaze over when I see the words "quantum fluctuation". I mean, I don't care that your maths say a universe can spontaneously arise out of nothing. It doesn't actually speak to the question why there is something rather than nothing. Hawking complains that if you say God did it, you simply shift the problem back to a First Cause. Well yes, but how is quantum did it any better?

I also have a lot of difficulty with inflation (not the thing with money although that also is a problem). A central problem for our cosmology is that the universe is too uniformly cool. By the second law of thermodynamics, heat transfers from the hotter to the cooler, but it needs time. The universe has not been around long enough for the even temperature we observe. So someone had the bright idea that in the universe's first second it exploded quicker than the speed of light for a short period, expanding enormously (I say enormously but it's all relative: I think it was something like the size of a grape after inflation). Okay, but what made it inflate? Time to get the hands waving because we have no idea. Some sort of repulsive force that only exists in very small, very hot balls of plasma. Which just happened to inflate the universe in precisely the way that leaves the universe we now observe.

Some of his other discussion is more successful. He clearly explains why we cannot have free will (which I've discussed before but I'm going to return to shortly because it's been on my mind). But he handwaves from that to a defence of strong AI, so that he claims that sufficiently complex computers will be self-aware. We don't know, and have no reason to believe, that's true. It might be, but it's a big claim that it is.

The thing is, there are facts, and there are opinions about the facts. Quantum theory is a good model because it explains how some of the facts work, but it's formal and interpreting it is to say the least difficult. It's also the case that the universe is consistent, so if you successfully explain some facts in your theory, you will tend to be explaining other similar facts. However, as Hawking to do him credit acknowledges, you are simply expressing a framework that fits, not saying what it is. The better our framework fits, the more we believe it is true, but the fact remains, it's just a framework.

1 Comments:

Blogger Don said...

Well, a giant can only reach so far from the shoulders he stands on.

Man I'm pithy.

November 12, 2010 at 6:48 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home